This study proposes a time-varying effect model that can be used to characterize gender-specific trajectories of health behaviors and conduct hypothesis testing for gender differences. The motivating examples demonstrate that the proposed model is applicable to not only multi-wave longitudinal studies but also short-term studies that involve intensive data collection. The simulation study shows that the accuracy of estimation of trajectory functions improves as the sample size and the number of time points increase. In terms of the performance of the hypothesis testing, the type I error rates are close to their corresponding significance levels under all combinations of sample size and number of time points. Furthermore, the power increases as the alternative hypothesis deviates more from the null hypothesis, and the rate of this increasing trend is higher when the sample size and the number of time points are larger.
BuuADabrowskaAMygrantsMet al.Gender differences in the developmental risk for onset of alcohol, nicotine, and marijuana use and the effects of nicotine and marijuana use on alcohol outcomes. J Stud Alcohol Drugs2014; 75: 850–858.
2.
BuuADabrowskaAHeinzeJEet al.Gender differences in the developmental trajectories of multiple substance use and the effect of nicotine and marijuana use on heavy drinking in a high risk sample. Addict Behav2015; 50: 6–12.
3.
WichstromL. The impact of pubertal timing on adolescents’ alcohol use. J Res Adolescence2001; 11: 131–150.
4.
Nolen-HoeksemaS. Gender differences in risk factors and consequences for alcohol use and problems. Clin Psychol Rev2004; 24: 981–1010.
5.
KeyesKMGrantBFHasinDS. Evidence for a closing gender gap in alcohol use, abuse, and dependence in the United States population. Drug Alcohol Depend2008; 93: 21–29.
6.
BodenJMFergussonDMHorwoodLJ. Associations between exposure to stressful life events and alcohol use disorder in a longitudinal birth cohort studied to age 30. Drug Alcohol Depend2014; 142: 154–160.
7.
SmithPHKaszaKAHylandAet al.Gender differences in medication use and cigarette smoking cessation: results from the international tobacco control four country survey. Nicotine Tob Res2015; 17: 463–472.
8.
ChenPJacobsonKC. Developmental trajectories of substance use from early adolescence to young adulthood: gender and racial/ethnic differences. J Adolesc Health2012; 50: 154–163.
9.
CaiZFanJLiR. Efficient estimation and inferences for varying-coefficient models. J Am Stat Assoc2000; 95: 888–902.
10.
DziakJJLiRZimmermanMAet al.Time-varying effect models for ordinal responses with applications in substance abuse research. Stat Med2014; 33: 5126–5137.
11.
QuAPLiR. Quadratic inference functions for varying-coefficient models with longitudinal data. Biometrics2006; 62: 379–391.
12.
TanXShiykoMLiRet al.A time varying effect model for Intensive longitudinal data. Psychol Meth2012; 17: 61–77.
13.
WangLBoKLiR. Local rank inference for varying coefficient models. J Am Stat Assoc2009; 104: 1631–1645.
14.
ZhuHLiRKongL. Multivariate varying coefficient model for functional responses. Ann Stat2012; 40: 2634–2666.
15.
LiuXLiRLanzaSTet al.Understanding the role of cessation fatigue in the smoking cessation process. Drug Alcohol Depend2013; 133: 548–555.
16.
SelyaASDierkerLCRoseJSet al.Time-varying effects of smoking quantity and nicotine dependence on adolescent smoking regularity. Drug and Alcohol Depend2013; 128: 230–237.
17.
VasilenkoSPiperMLanzaSTet al.Time-varying processes involved in smoking lapse in a randomized trial of smoking cessation therapies. Nicotine Tob Res2014; 16S2: S135–S143.
18.
YangHCranfordJALiRet al.Two-stage model for time-varying effects of discrete longitudinal covariates with applications in analysis of daily process data. Stat Med2015; 34: 571–581.
19.
ZuckerRAEllisDAFitzgeraldHEet al.Other evidence for at least two alcoholisms ii: life course variation in antisociality and heterogeneity of alcoholic outcome. Dev Psychopathol1996; 8: 831–848.
20.
Ryff CD and Almeida DM. National survey of Midlife in the United States (MIDUS II): Daily Stress Project, 2004-2009. ICPSR26841-v1. Ann Arbor, MI: Inter-university Cosortium for Political and Social Research [distributor], http://doi.org/10.3886/ICPSR26841.v1 (accessed 26 February 2010).
21.
McCullaghPNelderJA. Generalized linear models, London: Chapman and Hall, 1989.
22.
de BoorC. A practical guide to splines, New York, NY: Springer-Verlag, 1978.
23.
ShiykoMPLanzaSTTanXet al.Using the time-varying effect model (TVEM) to examine dynamic associations between negative affect and self-confidence on smoking urges: Differences between successful quitters and relapsers. Prevent Sci2012; 13: 288–299.
24.
SimpsonTLKivlahanDRBushKRet al.Telephone self-monitoring among alcohol use disorder patients in early recovery: a randomized study of feasibility and measurement reactivity. Drug Alcohol Depend2005; 79: 241–250.
25.
TuckerJABlumERXieLet al.Interactive voice response self-monitoring to assess risk behaviors in rural substance users living with HIV/AIDS. AIDS Behav2012; 16: 432–440.
26.
BartaWDTennenHLittMDMeasurement reactivity in diary research. In: MehlMRConnerTS (eds). Handbook of research methods for studying daily life, New York: Guilford Press, 2012, pp. 108–123.
27.
StritzkeWGKDandyJDurkinKet al.Use of interactive voice response (IVR) technology in health research with children. Behav Res Meth2005; 37: 119–126.
28.
BuuAJohnsonNJLiRet al.New variable selection methods for zero-inflated count data with applications to the substance abuse field. Stat Med2011; 30: 2326–2340.
29.
BuuALiRTanXet al.Statistical models for longitudinal zero-inflated count data with applications to the substance abuse field. Stat Med2012; 31: 4074–4086.
